Grounded within BPNT (Deci & Ryan, 2000), the current thesis comprised a series of

studies which aimed to develop and test an extended model of optimal and diminished human

functioning in sport, a life domain valued by many individuals across the world. The program

of research was designed primarily to provide unique empirical evidence to test the

propositions of SDT in relation to both the darker and brighter sides of human existence.

From an applied perspective, a greater, theoretically driven, insight into the mechanisms that

underpin indices of maladaptive and compromised functioning, as well as optimal well-being,

is important in order to support athletes realize their sport potential without hindering their

psychological and physical health.

Although a plethora of research has provided support for the hypothesized links

between autonomy support, need satisfaction, and psychological health and well-being (Adie,

Duda, & Ntoumanis, 2008; Gagné, Ryan, & Bargmann, 2003), the links between

interpersonal control, psychological need thwarting, and ill-being have been less frequently

studied within the SDT framework (Bartholomew, Ntoumanis, & Thøgersen-Ntoumani,

2010; Vallerand, Pelletier, & Koestner, 2008). This is primarily due to the way in which

interpersonal styles and psychological needs have been operationalized and measured in

research conducted to date. The current thesis aimed to address these measurement

limitations via the development and validation of two new psychometric scales: the

Controlling Coach Behaviors Scale (CCBS; Bartholomew et al., 2010; Chapter 2) and the

Psychological Need Thwarting Scale (PNTS; Bartholomew, Ntoumanis, Ryan, & Thøgersen-

Ntoumani, 2011; Chapter 3).

Subsequently, the findings from three further studies supported the application of a

model which incorporated assessments of athletes’ perceptions of coach control and

psychological need thwarting, alongside traditional measures of autonomy support and need

satisfaction, in the prediction of multiple indices of athlete well/ill-being (Chapter 4). In

particular, these findings supported the value of considering perceptions of controlling

interpersonal environments and feelings of psychological need thwarting in the prediction of

athlete ill-being.

Collectively, the results provided direct support for the propositions made by BPNT in

that optimal or compromised functioning depended upon the extent to which the three

psychological needs were perceived to be satisfied or thwarted in the sport environment (Deci

& Ryan, 2000). The present research is, however, the first to examine outcomes associated

with both the satisfaction and thwarting of needs simultaneously and in relation to both

supportive and controlling inputs within the same domain. The differentiated sets of findings

obtained in the three final studies have significant theoretical, measurement, and applied

implications for sport (and potentially other life domains). In this final chapter, the thesis

findings will be discussed in relation to these important advances and suggestions will be

presented for future research.

The Social Environment: Beyond Autonomy-Support

A key tenet of BPNT is that significant others (e.g., coaches) play a central role in

determining the quality of experience afforded to individuals in the setting in question via the

nature of the social environment they create (Ryan & Deci, 2007). Specifically, a coach’s

behavior can primarily be viewed in terms of two interpersonal styles: autonomy-supportive

and controlling. Research conducted to date has primarily focused on coaches’ autonomy-

supportive behaviors and the extent to which athletes experience psychological need

satisfaction (see Mageau & Vallerand, 2003). This thesis found consistent support for the

BPNT-based hypothesis that perceptions of autonomy-supportive social environments predict

the degree of need satisfaction experienced by athletes in the sport context.

 Beyond merely examining autonomy support, however, the findings from this thesis

also highlighted the importance of exploring the social-environmental conditions that

frustrate athletes’ psychological needs. To date, controlling coach behaviors have largely

been neglected in work examining social-environmental predictors of the three psychological

needs. In addition, when interpersonal control has been examined in relation to need related

outcomes, typically measures of satisfaction rather than thwarting have been used. The CCBS

(Bartholomew et al., 2010; Chapter 2) was designed explicitly to assess sports coaches’

controlling interpersonal styles from the perspective of SDT. Four separate controlling

motivational strategies salient in the context of sport were identified in the development and

initial validation of the measure: the controlling use of rewards, negative conditional regard,

intimidation, and excessive personal control. The scale demonstrated good content and

factorial validity, as well as internal consistency and invariance across gender and sport type.

However, further research is needed to test the temporal stability (test-retest reliability) of the

CCBS and its invariance across competitive level and experience. Subsequently, the

predictive validity of the measure in relation to multiple indices of ill-being was supported at

the between-person and within-person levels in Chapter 4.

Similarly to the majority of research grounded in SDT, the first two studies outlined in

Chapter 4 focused on predictors of between-person variability in need satisfaction and need

thwarting. In other words, these studies tapped athlete’s “general” perceptions of the degree

of autonomy support or control afforded in their training context. The third study extended

this work and demonstrated that situational perceptions of autonomy support and control

impact upon need satisfaction and need thwarting with immediate effect. Specifically, Study

3 found that the athletes’ perceptions of the interpersonal environment during training were

relevant to the degree of need satisfaction or need thwarting that the athletes reported

immediately after each session. The focus on within-person variability in perceptions of

controlling social environments created by significant others represents a particularly unique

aspect of this research and extends previous SDT-based diary research which has largely

neglected this important aspect of the social environment.

The findings from all three studies indicated that controlling strategies thwarted

athletes’ feelings of autonomy, competence, and relatedness. In turn, perceptions of need

thwarting were associated with the development of psychological and physiological ill-being

(Deci & Ryan, 2000). Thus, although controlling strategies can sometimes appear to be

adaptive in that they evoke desired behaviors and performance outcomes in the short term,

the current research suggests that such techniques may ultimately thwart athletes’

psychological needs and lead to diminished functioning.

At this juncture, it is important to distinguish between the use of control and the

provision of structure and involvement (Deci & Ryan, 1985; Mageau & Vallerand, 2003).

Structure refers to the provision and quality of information regarding expectations as well as

the provision of timely and useful feedback (Reeve, 2002). Involvement can be defined as the

extent to which a coach takes an active interest in and commits time, attention, and care

towards the athlete (Mageau & Vallerand, 2003; Gagné et al., 2003). Without instructions

and involvement from their coaches, athletes may lack the necessary information and

experience to interact competently with their environment. As long as the provision of

structure and involvement is not restrictive or intrusive (i.e., it must not be controlling;

Amorose, 2007) these contextual factors facilitate need satisfaction. The controlling strategies

identified in the CCBS clearly go beyond providing structure and positive emotional

involvement and attempt to manipulate athlete behavior and thus thwart psychological needs

with detrimental consequences for athlete well-being.

Alongside psychological needs and optimal/diminished functioning, it is also

important to explore the construct of interpersonal control in relation to the wider SDT
framework. Over the long term, continued exposure to controlling coach behaviors should not

only thwart athletes’ psychological needs, but also contribute to the development of

controlled motives (or amotivation). The current research is therefore limited in that this

proposition was not tested. Recent research using a version of the CCBS translated into

Spanish, however, has indicated that young football player’s perceptions of their coaches’

controlling behaviors are associated with amotivation and external and introjected regulation

and unrelated to identified regulation and intrinsic motivation (Castillo et al., 2010). Thus, the

controlling strategies identified in the CCBS are expected to have strong undermining effects,

perhaps even for athletes who have initially high levels of self-determination. This hypothesis

should be empirically tested in future longitudinal research. Although studies should explore

the moderating role of athlete resilience, we would suggest that most individuals subjected to

this kind of pressurized environment are likely to experience negative feelings that reflect

their needs being actively undermined by their coach.

Castillo et al. (2010) also presented psychometric evidence which supported the

factorial validity and internal reliability of the translated scale. It is important that cross-

cultural research continues to explore the nature and use of controlling coaching strategies in

sport. Due to the fact that the needs are hypothesized to be universal, the relations between

the psychological needs and well/ill-being should apply across cultures (Ryan & Deci, 2000).

However, the means through which the psychological needs are satisfied or thwarted may

vary as a function of culture. Thus, in an extreme case, it is possible for the same behavior to

be need satisfying for one group and need thwarting for another (Ryan & Deci 2002). This

suggests that there may be cultural differences in the way in which coach behaviors impact

upon athletes psychological needs. For example, recent research has shown that dominant-

submissive (i.e., controlling) interactions between coaches and athletes actually increased the

extent to which Chinese athletes experienced relationship satisfaction (Xin Yang & Jowett,
2010). To date, the CCBS has been translated and validated in Spanish, French, Greek,

Norwegian, Mandarin, and Flemish and findings from these studies are eagerly anticipated.

An overall measure of the controlling environment was used in each of the three

studies reported in the Chapter 4. This is because the primary focus of the current thesis was

on nomological relationships among higher order constructs (i.e., control and need thwarting;

relationships between specific sub-components of the two constructs should be the focus of

future work). As such, further research is needed to assess the utility of a differentiated

conception of controlling interpersonal behavior by exploring whether the separate

controlling strategies have individual effects on the three psychological needs, subsequent

motivation, and psychological and physical ill-being. It is likely that some maladaptive coach

behaviors may be more damaging to the motivation and well-being of athletes than others.

For example, the use of intimidation behaviors is expected to have a severe negative effect on

the well-being of those individuals subjected to them (Baker, Côté, & Hawes, 2000; Barber,

1996). Similarly, negative conditional regard has been associated with introjected regulation

and many serious forms of psychological ill-being in the parental literature (Hewitt & Flett,

1991). Thus, controlling motivational strategies which attempt to control athlete behavior by

overtly manipulating or exploiting the coach-athlete relationship (through intimidation and

negative conditional regard) may produce the most damaging effects upon athlete motivation

and well-being, when compared to strategies such as the use of tangible rewards or

controlling praise.

In contrast to the other facets of controlling behavior incorporated within the CCBS,

the controlling use of rewards is an approach-oriented strategy in the sense that rewards and

praise induce hopes for increased recognition, attention, and self-aggrandizement. Thus,

future research exploring reward and praise may be of particular importance because the

current findings indicated that approach-oriented, and apparently more benign control, can
also be problematic. Recent research by Roth, Assor, Niemiec, Ryan, and Deci (2009)

supported this proposition. Compared to more avoidance-orientated means of control, the use

of positive conditional regard – a strategy in which parents motivate children to comply by

providing more regard (affection, recognition, and attention) when children behave in

accordance with parental expectations – was associated with fragile and unstable self-esteem,

which drove children to over-strive and compulsively invest effort in the enactment of

parental valued behaviors (i.e., introjected motivation aimed at attaining positive regard). In

contrast, Roth et al. suggested that the use of more punitive means of control (i.e., avoidance-

orientated strategies such as negative conditional regard and intimidation) are more likely to

be associated with amotivation and depression (Roth et al., 2009). Thus, the distinction

between the various components of controlling coaching and their differential impacts upon

psychological needs, motivation, and well-being is an interesting avenue for future research.

In line with previous research (Pelletier, Fortier, Vallerand, & Brière, 2001; Silk,

Morris, Kanaya, & Steinberg, 2003; Tessier, Sarrazin, & Ntoumanis, 2008), the findings from

this thesis indicate that the support of autonomy and the control of behavior are not two sides

of the same coin. As such, coaches may use a variety of autonomy-supportive and controlling

behaviors simultaneously and to a different extent. Thus, an absence of autonomy support

does not necessarily equate to high levels of control (Bartholomew, Ntoumanis, &

Thøgersen-Ntoumani, 2009) and coaches can also use neutral strategies (Tessier et al., 2008).

Incorporating perceptions of controlling coach behaviors alongside autonomy support,

therefore, reflects a more comprehensive examination of the diverse features of the social

environment which impact upon athletes’ experiences of need satisfaction and need

thwarting. In summary, considering these different facets of the social environment, which

past research has suggested are related but distinct constructs (Pelletier et al., 2001; Silk et
al., 2003), should allow better understanding of the variability in psychological experiences

of individuals in sport, and in other life domains.

The Three Basic Psychological Needs: Beyond Need Satisfaction

The role of psychological need satisfaction/thwarting in the ‘social environment –

well/ill-being’ dialectic is the focus of BPNT (Ryan & Deci, 2000). To qualify as a need,

therefore, a motivating force must have a direct relation to well/ill-being (Ryan & Deci

2002). The proposition that psychological needs, when satisfied, promote athlete well-being

was fully supported in the current research, both at the between-person and within-person

level (Chapter 4). However, the current thesis aimed to extend previous research conducted in

the sport domain by explicitly assessing need thwarting in relation to ill-being outcomes. The

findings presented in Chapters 3 and 4 demonstrated that need thwarting has meaningful and

independent empirical consequences beyond more traditional measures of need satisfaction.

Such findings should allow researchers to provide better predictions regarding variability in

the negative psychological and physiological health-related outcomes experienced by athletes

in sport and provide new empirical support for the role of need thwarting in SDT’s theoretical

account of the darker sides of human existence.

A number of limitations relating to the way in which need thwarting had been

conceptualized and measured in previous questionnaire-based research were highlighted

during the development of the PNTS (Bartholomew et al., 2010; Chapter 3). These included

the positive conceptualization of the three psychological needs and subsequent failure of

assessment tools to take into account the active nature and intensity of need frustration that

characterize states of need thwarting (Deci & Ryan, 2000). Specifically, need thwarting does

not simply reflect the perception that need satisfaction is low, but moreover the perception

that need satisfaction is being obstructed or actively frustrated within a given context. In

addition, need thwarting items must capture the intensity of the negative experiential state
which occurs when an individual’s psychological needs are actively frustrated (i.e., one feels

oppressed, inadequate, and rejected).

As a result of the measurement limitations outlined above, previous research

exploring negative relations between need satisfaction and maladaptive outcomes has only

provided indirect evidence to support the detrimental effects of need thwarting on athlete

health and well-being (e.g., Lonsdale, Hodge, & Ng, 2008; Reinboth, Duda, & Ntoumanis,

2004). In addition, because previous research has not distinguished between low levels of

need satisfaction and need thwarting, the hypothesized relations between the psychological

needs and indices of ill-being have not always been supported in the literature (Adie et al.,

2008; Gagné et al., 2003; Quested & Duda, 2010). That is, the psychological needs do not

always relate significantly (and negatively) to indicators of ill-being (e.g., negative affect:

Gagné et al., 2003; McDonough & Crocker, 2007; and burnout symptoms: Quested & Duda,

2010). Thus, the current research aimed to provide direct support for the propositions

outlined by BPNT in relation to the darker side of sport participation by examining the

relationship between explicit measures of need thwarting and diminished functioning and ill-

being.

The PNTS is a multidimensional measure designed to tap the negative experiential

state which occurs when athletes’ perceive their psychological needs for autonomy,

competence, and relatedness to be actively undermined in the sport environment. Similarly to

the CCBS, future research is required to test the temporal stability of the scale. Overall,

however, the scale demonstrated good content, factorial, and predictive validity, as well as

internal consistency and invariance across gender, sport type, competitive level, and

competitive experience. The development of a valid and reliable need thwarting measure was

the first step in the exploration of this relatively understudied but theoretically important

construct.
 The data presented in this thesis provided consistent support for the assessment of

need thwarting independently of need satisfaction. In Chapter 3, small negative correlations

were observed between athletes’ perceptions of need thwarting and need satisfaction

indicating that, empirically, need thwarting and need satisfaction may not be antipodal.

Further, the results of three separate EFA analyses showed that, within each need, need

thwarting and need satisfaction represented distinct factors. These findings suggest that

perceived need satisfaction and perceived need thwarting are independent constructs. There

are, however, problems inherent with the use of factor analysis in this regard. For example,

when a set of items that are indicators of opposite poles of a single underlying bipolar

dimension are submitted to factor analysis, two factors generally appear instead of one factor:

one factor including the positive items and one factor including the negative items (González-

Romá, Schaufeli, Bakker, & Lloret, 2006; Lloret & González-Romá, 2000). This is because

data gathered by means of balanced Likert-type scales composed of positive and negative

items do not fit the linear factor analysis model (Van Schuur & Kiers, 1994). An interesting

avenue for future research, therefore, would be to conduct Item Response Theory (IRT)

analyses (which do not assume linear relationships) to further examine whether or not need

satisfaction and need thwarting are scalable on a single continuum (Edelen & Reeve, 2007).

In line with BPNT, Chapter 3 presented preliminary evidence to suggest that the

PNTS yielded better predictions concerning negative outcomes associated with sport

participation (i.e., emotional and physical exhaustion) compared to existing measures of

psychological need satisfaction. The regression analyses clearly demonstrated the value of

considering perceptions of need thwarting. Within each need, need thwarting accounted for

additional variance above and beyond that due to need satisfaction scores. Further, in the

subsequent SEM analysis, emotional and physical exhaustion was predicted by need

thwarting only. These initial findings were further supported in Chapter 4. Cross-sectional
and diary evidence from three independent samples suggested that the manifestations of ill-

being in sport (i.e., disordered eating, burnout, depression, negative affect, symptomatology,

and perturbed physiological functioning) were more related to the presence of need thwarting

than to the absence of need satisfaction. The finding that these dimensions differently

predicted mental health outcomes supplies further impetus for disentangling these constructs

in future SDT-based research. In particular, future experimental work on need thwarting

would complement the current studies.

As well as examining outcomes associated with both the satisfaction and thwarting of

needs simultaneously, the three studies conducted in Chapter 4 examined both supportive and

controlling inputs within the same domain. Need satisfaction was predicted by athletes’

perceptions of coach autonomy-support and need thwarting was better predicted by coach

control. These findings further supported the notion that need satisfaction and need thwarting

are best viewed as independent constructs which not only predict different outcomes, but

have separate antecedents. This provides unique empirical evidence to support the processes

outlined by BPNT in relation to the darker and brighter sides of human functioning.

Given their potential independence, it is plausible that alongside perceived need

satisfaction, athletes can also experience the active thwarting of needs within the same

environment. For instance, research suggests that coaches can simultaneously engage in

behaviors that support and forestall athletes’ feelings of autonomy, competence, and

relatedness (Pelletier et al., 2001). In line with observational research on coaching

environments (Smoll & Smith, 2002), the findings presented in Chapter 3 suggested that

perceptions of autonomy and competence need satisfaction and need thwarting could

sometimes co-occur. The interactions observed in Chapter 3 generally indicated that

buffering effects can be observed between corresponding need satisfaction and need

thwarting constructs. For example, the deleterious effects of need thwarting on well-being
may be moderated via athletes’ perceptions of need satisfaction. Although similar interactions

were not examined in Chapter 4, future research should further explore these opposing, yet

sometimes co-occurring, dynamics within the sport context.

In order to explore nomological relationships, an overall measure of need thwarting

was used in each of the studies reported in Chapter 4. It would, however, be interesting to

explore whether it is necessary for all three needs to be thwarted in order for ill-being to

occur. Thus, another line of research would be to explore the individual contributions of each

of the three needs in the etiology of diminished functioning and ill-being. Deci and Ryan

(2000) suggested that specific patterns of thwarted need satisfaction may be key

psychological contributors to many mental disorders and their behavioral manifestations.

Future studies may, therefore, wish to employ the three-factor PNTS model, supported in

Chapter 3, to investigate whether the thwarting of one need is more strongly linked to

particular negative outcomes. Although the three psychological needs tend to be highly

correlated and function in unison in natural settings (e.g., Baard, Deci, & Ryan, 2004; Gagné

et al., 2003), previous research has observed differences in the extent to which the three

psychological needs account for levels of psychological well/ill-being (e.g., Adie et al., 2008;

McDonough & Crocker, 2007). In research undertaken with athletes, competence has been

shown to be the most salient need in predicting targeted motivational and well/ill-being

outcomes (e.g., Reinboth et al., 2004). Such findings suggest that the functional significance

of the situation can influence the relative impact of each need upon cognitive, behavioral, and

affective outcomes (Deci & Ryan, 1985). Clearly, the demonstration of competence is

important in competitive sport and thus, the thwarting of this particular need may make a key

contribution to athlete ill-being.

It is also important for future research to explore need thwarting in relation to the

wider SDT framework. For example, future research should examine the direct effects of
psychological need thwarting on the development of the three defensive psychological

accommodations assumed to have severe costs for mental health and well-being: the

development of controlling regulatory styles, compensatory motives, and rigid behavior

patterns (Deci & Ryan, 2000). Of particular theoretical importance is the exploration of need

thwarting in relation to motivation (i.e., controlling regulatory styles). Need thwarting is

hypothesized to have an undermining effect on self-determination and promote controlled

forms of motivation such as external and introjected regulation. However, this assumption

was not tested in the current thesis.

Compensatory motives are need substitutes (e.g., extrinsic goals) that do not really

satisfy the thwarted need but provide some collateral satisfaction (Deci, 1980). Future

research is needed to explore the impact of need thwarting on the development of extrinsic

goals in sport. Initial support for the role of need thwarting in the development of rigid

behavior patterns (i.e., disordered eating) was provided in Chapter 4. Although rigid behavior

patterns help protect individuals from the inner hurt that results from psychological need

thwarting, they also tend to prevent individuals from dealing with their inner feelings and

experiencing subsequent need satisfaction (Deci & Ryan, 2000). It would be interesting,

therefore, for future research to extend this preliminary work by examining need thwarting

and rigid eating patterns from a longitudinal perspective in order to explore whether

psychological accommodations lead to further thwarting of need satisfaction over time.

Recent research has also suggested that unmet psychological needs can increase

corresponding desires to seek need fulfillment experiences (Moller, Deci, & Elliot, 2010;

Sheldon & Gunz, 2009). Such findings suggest that athletes experiencing high levels of need

frustration in the sport environment may be motivated to increase their feelings of autonomy,

competence, and relatedness in sport and/or in other life contexts (Vallerand, 2000).

However, it is unlikely that individuals will always respond in such a positive fashion to need
thwarting. It may be that when individuals are thwarted, they are just undermined and

alienated (as opposed to spurred into action) and, therefore, turn to less optimal ways of

functioning. It would be interesting for future research to explore the circumstances under

which athletes who experience need thwarting attempt to seek need fulfillment experiences

within their sport, in other life contexts, or by turning to need substitutes and other less

optimal ways of functioning. The dynamics of subsequent motivation after need thwarting are

clearly complex, rather than straightforward, and demand future research attention.

Finally, although the means through which needs are satisfied or thwarted may vary

as a function of culture, as mentioned previously, the underlying processes via which the

psychological needs promote or forestall health and well-being are theorized to be the same

across all groups (Ryan & Deci, 2002). Accordingly, future research is needed to explore the

construct of need thwarting from a cross-cultural perspective. Recent research incorporating a

Spanish version of the PNTS has provided initial cross-cultural evidence to support the role

of need thwarting in athlete ill-being by demonstrating that the thwarting of each

psychological need (as assessed using a translated version of the PNTS) was significantly

related to emotional and physical exhaustion in a sample of young Spanish football players

(Balaguer et al., 2010). A number of additional findings also supported the factorial validity

and internal reliability of the Spanish PNTS. Similarly to the CCBS, the PNTS has been

validated in several languages for use in sport (or adapted to be used in other life domains)

and further findings relating to the process of need thwarting from a cross-cultural

perspective are anticipated.

In summary, the findings of this thesis support the notion that psychological need

thwarting is not equivalent to low levels of need satisfaction. Need satisfaction and need

thwarting are, therefore, best viewed as independent constructs. As expected, the assessment

of need thwarting added significantly to the prediction of diminished functioning and ill-
being. Thus, the current research provided new support for the assumptions made by BPNT

in that both perceptions of psychological need satisfaction and need thwarting simultaneously

impacted upon athletes’ functioning and explicit assessments of both more fully addressed

the multiple impacts of sport participation on wellness and ill-being.

Athlete Well/Ill-being: A Function of the Social Environment and Psychological Needs

Although sport participation is generally associated with positive outcomes including

increased psycho-social development and physical health (Fraser-Thomas, Côté, & Deaken,

2005), there is evidence to suggest that athletes may be ‘at risk’ of ill-being and particular

psychological and physical problems (Gould, 1993; Theberge, 2008; Sundgot-Borgen &

Torstveit, 2004). SDT principally embraces the eudaimonic conception of well-being and

examines the processes that underlie healthy, congruent, and vital functioning (Ryan, Huta, &

Deci, 2008). In line with the conceptualization of well-being endorsed by SDT, it is also

important to recognize that well-being and ill-being are not antipodal (Ryan & Deci, 2001).

Thus, the absence of psychological or physical ill-health does not necessarily equate to

optimal functioning, and vice versa. Therefore, a range of outcomes were targeted in the

current thesis to capture the processes and experiences associated with optimal (i.e., vitality

and positive affect) and, in particular, diminished functioning (negative affect, disordered

eating, burnout, physical symptoms, and perturbed physiological arousal).

The descriptive statistics reported in this thesis indicated that in general, athletes

experienced low levels of depression and negative affect and moderate burnout symptoms

(mean scores on the burnout measure were just under the midpoint). However, over one-

quarter of the female athletes involved in the first study reported in Chapter 4 were identified

as either displaying symptoms of disordered eating or as having a clinical eating disorder.

This is in line with research suggesting that sport participation is associated with an increased
incidence of subclinical eating problems and/or eating disorders (Sundgot-Borgen, 1993). In

addition, whilst, on average, athletes experienced high levels of vitality, the mean for positive

affect was only marginally above the scale’s midpoint. Collectively, these findings support

the proposition that healthful sport participation is not automatic (Wankel & Mummery,

1990). Thus, in order to promote adaptive sport participation, it is important to understand the

factors which contribute to the variability in health-related outcomes.

The descriptive statistics reported in Chapter 4 indicated that, on average, athletes

reported high levels of autonomy support and need satisfaction. This suggests that many

coaches exhibited adaptive behaviors that facilitated the satisfaction of their athletes’

psychological needs. In addition, relatively low mean scores suggested that most athletes did

not perceive their coaches to be overly controlling or experience high levels of need

thwarting in the sport context. This is seemingly good news for those interested in supporting

athlete welfare and promoting positive experiences in sport. However, this picture must be

interpreted with caution. Firstly, in all studies, there was evidence of variability in each of the

targeted variables. This indicates that some athletes did have negative experiences in the sport

context, report high levels of need thwarting, and/or experience psychological and physical

ill-being. Secondly, it is important to acknowledge that the coaches’ agreement to participate

in this research may have reflected an existing interest in creating adaptive coaching

environments. Controlling coaches are more likely to be protective not only of their athletes,

but also of their own coaching philosophy and the training environments they create.

Therefore, the thesis data may have been more variable and we may have observed higher

mean levels of control, need thwarting, and athlete ill-being, if consent could have been

obtained from a more representative sample of coaches in the UK.

In line with previous research, the findings from the current thesis confirmed the

utility of examining optimal athlete well-being from a need-fulfillment perspective (Adie et

al., 2008; Gagné et al., 2003; Reinboth et al., 2004). For example, vitality and positive affect

were employed to gauge optimal functioning and emotional well-being in the three studies

reported in Chapter 4. At the between-person level, need satisfaction positively predicted

athletes’ typical feelings of vitality and experiences of positive affect (Studies 1 and 2,

respectively). However, SDT also postulates that need satisfaction should predict enduring

signs of effective functioning and psychological health (Deci & Ryan, 2000). The diary

technique afforded the possibility of exploring the social-psychological processes that

underlie within-person variability in reported well-being over time (Bolger, Davis, & Rafaeli,

2003). Via the application of this method (Study 3), daily experiences of need satisfaction

during training were shown to predict athletes’ daily experiences of positive affect. This adds

weight to the BPNT postulate with regard to the psychological mechanisms important for

sustained well-being. The social-psychological processes underlying well-being, however,

were not the primary focus of the present research.

If the needs for autonomy, competence, and relatedness are thwarted, ill-being and

compromised functioning are expected to ensue (Deci & Ryan, 2000). However, the

conditions and processes that lead to the manifestation of ill-being have received less

attention in the SDT literature. Overall, the findings from the current thesis support the notion

that perceptions of interpersonal control and subsequent need thwarting contribute to the

manifestation of compromised functioning and ill-being. Multiple indices of ill-being were

targeted in Chapter 4 including disordered eating, which was predicted by athletes’

perceptions of psychological need thwarting only (Study 1). Thus, it would appear that the

struggle for body control can represent a compensatory process prompted by frustrations in

perceived autonomy, competence, and relatedness needs. Similarly, the experiential state of

burnout was more related to the active thwarting of psychological needs than to a perceived

lack of need satisfaction (Study 2). These studies also indicated that general perceptions of
need thwarting were associated with between-person variability in depression and negative

affect. Study 3 aimed to examine the emotional impact of need thwarting in more detail by

exploring changes in daily levels of negative affect from before to after training using the

diary methodology. Physical symptoms were also measured in order to explore the proposed

link between psychological needs and self-reported physical health. Perceptions of

psychological need thwarting during training predicted changes in negative affect and

physical symptoms from before to after training. Specifically, after controlling for pre-

training levels, athletes experienced more post-training negative affect and physical

symptoms on the days in which they reported higher levels of perceived need thwarting

during training. The fact that the same processes were observed at the between and within-

person levels provides strong support for the utility of considering interpersonal control and

need thwarting in the prediction of diminished functioning and ill-being.

This thesis also aimed to further extend prior research by including a biological

marker responsive to psychological stress. Secretory immunoglobulin A (S-IgA) is an

immunological protein. Its main purpose is to protect against the invasion of infectious agents

(e.g., viruses, bacteria). Acute psychological stressors, with the exception of stressors

involving physical pain, have consistently been found to increase S-IgA levels (Bosch, Ring,

de Geus, Veerman, & Amerongen, 2002). Aligned with the predictions of BPNT, the findings

from Study 3 indicated that athletes who generally experienced need thwarting in the sport

environment were more susceptible to biological and psychological deregulation.

Specifically, the results indicated that athletes who perceived their needs to be actively

thwarted in the sport environment were more likely to show enhanced physiological arousal

(i.e., elevated levels of S-IgA) immediately prior to training, possibly reflective of

anticipatory apprehension.
 These findings make a novel and important contribution to the literature and support

SDT as a relevant framework within which to advance our understanding of the motivational

factors that may contribute towards psychophysiological stress responses. The study of

biological mechanisms is underexplored within SDT, and could easily be expanded to include

other markers of immunological functioning known to be associated with stress, such as the

hormone Cortisol. Repeated exposure to situations perceived as excessively stressful may

have long term physical (Burns, 2006) and psychological (Raedeke & Smith, 2004) health

implications. Thus, the inclusion of biomarkers (e.g., S-IgA, Cortisol) in future longitudinal

SDT-based research should be particularly informative regarding the potential immunological

pathways by which social-psychological processes lead to variability in health (and ill-

health).

It would also be interesting for future research to explore the interdependencies

between interpersonal control, need thwarting, state emotional responses, and indices of

immunological functioning. In the current research, positive and negative affect were

measured at a global level only and thus did not correlate significantly with S-IgA (indicative

of physiological arousal and acute stress immediately prior to training). Moreover, it may

also be that state-level indicators of need thwarting produce even stronger associations with

biological markers responsive to acute stress. Biological stress responses (i.e., elevated

salivary Cortisol levels) have also been associated with exposure to controlling interpersonal

behaviors during a learning activity in the educational setting (Reeve & Tseng, 2010).

Future research could also provide further support for the construct of need thwarting

by exploring the distinct biological consequences of need satisfaction and need thwarting. In

line with the conceptualization of well-being endorsed by SDT, recent biological research has

suggested that the biomarkers of well-being and ill-being are primarily ‘distinct’ (i.e., well-

being and ill-being have different biological signatures; Ryff et al., 2006). Based on the
theoretical hypotheses underlying the current thesis, perceptions of need satisfaction should

predict biological correlates of well-being and perceptions of need thwarting should predict

biological correlates of ill-being. Such an approach highlights the benefits that could be

accrued from a multi-disciplinary approach to the study of the social-psychological predictors

of ill-being and would provide new and stronger support for assessing need thwarting

independently of need satisfaction.

Practical Implications

Grounded within BPNT, this thesis was undertaken with the intention of addressing

the darker side of sport participation and the negative health-related implications of

interpersonal control and psychological need thwarting. Aside from the desire to make a

significant theoretical and empirical contribution to the SDT literature, this research was

motivated by a personal aspiration to contribute towards the advancement of training

environments which help athletes realize their sport potential without compromising their

health and well-being. The practical implications stemming from this thesis are thus aligned

with the conceptual foundation upon which the studies were based.

Overall, the findings supported the propositions made by BPNT and indicated that the

social environment in which training is carried out has an important influence on athlete

well/ill-being. In particular, the different instructional environments manifested in sport via

the interpersonal style of the coach clearly play a pivotal role in supporting or thwarting

athletes’ psychological needs and shaping subsequent health-related outcomes. The fact that

the differing interpersonal behaviors employed by coaches can have such a profound effect

upon the psychological and physical health of athletes has significant applied ramifications.

In relation to the brighter side of sport participation, the findings corroborated previous

research (Adie et al., 2008; Gagné et al., 2003; Reinboth et al., 2004) and indicated that
autonomy-supportive training environments supported athletes’ psychological needs and

helped them experience vitality and sustained positive emotions. Research in exercise and

educational settings has indicated that it is possible to train instructors and teachers to be

more autonomy-supportive (Edmunds, Ntoumanis, & Duda, 2008; Reeve, 1998; Tessier,

Sarrazin, & Ntoumanis, 2010). Evidence-based interventions such as these provide authority

figures with the skills to be more autonomy-supportive and help promote positive outcomes

and well-being for individuals within the setting in question.

Beyond training individuals to be more autonomy-supportive, however, coaches must

also be equipped with the skills to identify and avoid the use of controlling interpersonal

behaviors (e.g., Tessier et al., 2008). The current thesis identified four/five maladaptive

motivational strategies associated with a controlling interpersonal style in sport: the

controlling use of rewards, negative conditional regard, intimidation, excessive personal

control, and behaviors which judge or devalue the athlete. These strategies were well aligned

with the SDT literature and perceived to occur frequently in the sport domain (Chapter 2).

Subsequent findings indicated that controlling training environments thwarted athletes’

psychological needs and were associated with negative feelings and maladaptive behavioral

outcomes (Chapter 4).

It is, therefore, important to understand why coaches engage in strategies that have the

potential to be psychologically damaging for their athletes in the first place. Research

undertaken in educational settings indicates that there are several factors which influence the

extent to which a teacher adopts a controlling interpersonal style. To organize these multiple

influences into a coherent framework, it is important to distinguish between “pressures from

above” (e.g., administrators), “pressures from below” (e.g., students), and “pressure from

within” (e.g., teachers own values and personality depositions; Pelletier, Seguin-Levesque, &

Legault, 2002; Reeve, 2009). When teachers are themselves pressured to produce particular
student outcomes (e.g., high grades, good behavior) they are more likely to teach in a

controlling fashion (Flink, Boggiano, & Barrett, 1990). Similarly, when teaches perceive their

students to be disruptive or low in motivation and engagement they tend to react by adjusting

their behavior toward a more controlling style (Sarrazin, Tessier, Pelletier, Trouilloud, &

Chanal, 2006). In addition, research suggests that some teachers are motivationally or

dispositionally orientated toward a controlling style (Pelletier et al., 2002). When teachers

enter the classroom with controlled motivation of their own and harbor controlling

orientations within their personality (e.g., are highly authoritarian or conservative) they are

more likely to adopt a controlling style toward students (Cai, Reeve, & Robinson, 2002).

Research conducted in the sport and physical education literature has also identified a

number of demands on coaches which include pressures from above (e.g., external

evaluations of athlete performance, time constraints, and a lack of assistance and guidelines

from sport organizations), pressures from below (e.g., perceived athlete motivation), and

pressures from within (e.g., coach personality; Allen & Shaw, 2009; Stebbings, Taylor, &

Spray, in press; Taylor, Ntoumanis, & Standage, 2008). In addition, recent research in sport

has pointed to the importance of coaching contexts that facilitate coaches’ psychological need

satisfaction and well-being. For example, Stebbings et al. (in press) found that competence

and autonomy need satisfaction predicted coaches’ levels of psychological well-being (as

indexed by positive affect and subjective vitality) which, in turn, negatively predicted their

perceived controlling behaviors towards athletes. Thus, it is likely that coaches will adopt

more positive coaching behaviors when their own needs are satisfied and more controlling

behaviors when their own needs are thwarted. An exploration of the pressures that thwart

psychological needs and result in a greater reliance on maladaptive coaching strategies is

clearly warranted. It would, therefore, be interesting to explore how an explicit measure of

coaches’ experiences of need thwarting relates to their use of controlling behaviors.

Moreover, athletes’ perceptions of coach behavior would complement coaches’ self-reports,

especially as athletes’ perceptions of the social environment are most pertinent in predicting

athlete consequences (Deci & Ryan, 1987).

Given the pressure that some coaches are under to produce results, their adoption of

more controlling behaviors towards their athletes is understandable. However, it is clear that

both coaches and athletes function better when coaches support athletes’ autonomy (Deci, La

Guardia, Moller, Scheiner, & Ryan, 2006; Gagné et al., 2003; Stebbings et al., in press).

Thus, coaches must try to become less controlling (i.e., avoid controlling sentiment,

controlling language, and controlling behaviors; Mageau & Vallerand, 2003; Reeve, 2009).

Identifying the factors which promote the adoption of controlling behaviors is important so

that this awareness can allow coaches to become more mindful of the forces that take them

away from supporting athletes’ autonomy. Similarly, a greater understanding of the

detrimental impact of controlling coaching strategies should help coaches become more

aware of the inimical effect their behavior can have on athletes. Therefore, as coaches

become more mindful of the causes and consequences of their controlling behaviors they

should gain a greater capacity to self-reflect upon the motivational strategies they employ and

this should facilitate the adoption of more adaptive coaching behaviors (e.g., Reeve, 2009).

Drawing from the educational literature, it is apparent that coaches must volitionally

endorse the practice of an autonomy-supportive style (i.e., they must want to support their

athletes autonomy). They must, therefore, deeply appreciate the benefits of such action and

create conditions that enable the practice of an autonomy-supportive style to take root (i.e.,

take the athletes’ perspective, welcome athletes’ thoughts, feelings, and behaviors, and

support athletes’ capacity for autonomous self-regulation). The next task in trying to become

more autonomy-supportive is to become aware of, develop, and ultimately refine the

interpersonal behaviors that actualize an autonomy-supportive style (Reeve, 2009). However,

autonomy support is not just a list of skills or behaviors and identifying exactly how to help

coaches adopt a more autonomy-supportive approach, regardless of the situational demands,

remains a challenge for sport psychologists.

Limitations and Future Directions

A number of limitations and ideas for future research have been discussed throughout

this chapter and will be summarized in this penultimate section. The primary limitation of the

current research is its focus on BPNT, which can be viewed as somewhat narrow. The

constructs of interpersonal control and psychological need thwarting still need to be

incorporated within the wider SDT framework. In this regard, it is important that future

research assesses athlete motivation alongside interpersonal control and need thwarting.

Specifically, chronic exposure to controlling interpersonal environments and experiences of

need thwarting should lead to controlled motives/amotivation (as well as other psychological

accommodations including need substitutes and the development of rigid behavior patterns).

It would also be interesting to explore the moderating role of athlete resilience and

subsequent motivation after need thwarting (i.e., do thwarted needs motivate subsequent

desires for need fulfillment experiences or do individuals just turn to less optimal ways of

functioning?).

Thus, longitudinal research examining the dynamic interplay between motivational

constructs (i.e., autonomy-support, control, need satisfaction, need thwarting, and

motivational regulations) would clearly be of great value. The findings presented in the

current thesis suggest that perceptions of need satisfaction and need thwarting can co-exist.

Thus, more research is needed to explore the dynamic interplay between these opposing yet

co-occurring experiences. It could be that athletes who experience high levels of need

thwarting and high levels of need satisfaction in the sport environment are less vulnerable

than athletes who experience high levels of need thwarting and low levels of need
satisfaction. In addition, at the momentary level, it may be possible for athletes to

simultaneously experience the satisfaction of one need and the thwarting of another need. The

exploration of these dynamics in relation to motivation and well/ill-being outcome variables

(e.g., via diary studies) could be particularly informative. Research could also explore

interactions between perceptions of coach autonomy-support and control and assess their

subsequent impact on experiences of need satisfaction and need thwarting. For example, the

use of controlling strategies may not be as detrimental to experiences of autonomy,

competence, and relatedness if they are used alongside more autonomy-supportive behaviors.

These motivational dynamics should also be explored from a cross-cultural perspective (e.g.,

the nature, use, and impact of controlling coaching strategies on psychological needs,

motivation, and well/ill-being in authoritarian cultures such as China). In addition, research

exploring the individual contributions of the separate controlling strategies identified in the

CCBS, and the individual role of each thwarted need, in the etiology of diminished

functioning and ill-being should also prove profitable.

Despite including an objective immunological marker in Chapter 4, the current thesis

relied heavily on self-report measures. Whilst Deci and Ryan (1984) have suggested that it is

the subjective perception and interpretation of social-environmental factors which influence

ones affective state and behavior, objective measures of the sport environment (e.g., coach

behavioral observations) and athlete well/ill-being (e.g., biological stress markers) should be

employed alongside subjective assessments in future work. In this regard, research should

continue to explore psychophysiological stress responses and immunological functioning

within the SDT framework.

The current thesis focused primarily on health-related outcomes and well/ill-being.

Another important avenue for future research, therefore, is the exploration of interpersonal
control and need thwarting in relation to additional outcomes associated with sport

participation including athletic performance and other aspects of psycho-social development.

Finally, it is also important that the practical implications stemming from these

findings [i.e., coaches must be equipped with the skills to (a) identify and avoid the use of

controlling interpersonal behaviors and (b) adopt more autonomy-supportive strategies] are

put into practice. Thus, interventions which help coaches foster quality motivation and athlete

enjoyment are needed to ensure that athletes are able to realize their potential without

compromising their health and well-being.

Summary and Conclusion

Throughout this final chapter, the most pertinent findings from the thesis have been

presented and discussed. Collectively, the findings supported a new empirical approach to

understanding diminished functioning and ill-being in sport. Firstly, exploring the social-

environmental conditions that sustain and frustrate athletes’ psychological needs reflected a

more comprehensive examination of the social environment (compared to traditional

approaches) and clearly demonstrated the importance of measuring autonomy-supportive and

controlling behaviors independently. Secondly, the current research was the first to

empirically distinguish between low levels of need satisfaction and need thwarting. The two

independent need constructs not only had separate antecedents, but also predicted different

outcomes and even co-occurred in the same setting. Finally, the manifestation of ill-being in

sport was consistently more related to the presence of need thwarting than to the absence of

need satisfaction. This highlights the importance of incorporating direct assessments of need

thwarting when diminished psychological and physiological functioning is the focus of an

investigation.
 Taken in their totality, these findings provide new empirical support for the role of

interpersonal control and need thwarting in SDT’s theoretical account of the darker sides of

human existence and should allow researchers to provide better predictions regarding

variability in negative psychological and physiological health-related outcomes. The ongoing

application of the SDT framework as an approach to understanding compromised

functioning, as well as optimal well-being, will be important in addressing both the darker

and brighter sides of sport participation and helping athletes realize their sport potential

without compromising their health and well-being. Controlling interpersonal behaviors and

psychological need thwarting must be further researched if the development of diminished

functioning and ill-being is to be properly understood and prevented. Clearly, this issue is not

only relevant to the sport domain and, therefore, the conceptual and measurement

implications of the current findings should also be studied in other social contexts including

education, work, parenting, health-care, psychotherapy, and personal relationships.